The present invention relates generally to electricity theft detection and, more particularly, to a system and method of detecting theft of electricity by performing an integrity checks analysis on a power distribution system using electrical data from various metering devices.
Theft of electricity is a serious problem worldwide. Electricity theft has become the third largest form of theft behind credit card data theft and automobile theft. In 2014, worldwide losses as a result of stolen electricity amounted to 89.3 billion dollars. Electricity theft losses in the United States amount to approximately 6 billion dollars every year. Approximately 80% of electricity theft is residential, with the remaining 20% of electricity theft being commercial. Electricity theft is one of the most prominent, if not the most prominent, form of non-technical losses. Non-technical losses are caused by actions external to a utility's power distribution system or caused by loads and/or conditions not taken into account in the computations for the power distribution system technical losses (naturally occurring or internal losses from power dissipation).
A variety of methods are used by utility customers to steal electricity from electric utilities. Many of the methods involve tampering with the primary meter that reads the electricity flowing into the residential or commercial load. One way to tamper with older meters is to pull out the meter that connects the electrical path from the utility to the electrical path into the property and put that meter back in upside down. Thus, the line side of the meter and the load side of the meter would be reversed, and the meter would record any measurements taken as a reverse flow of electricity. In other words, the meter would read that electricity is being provided to the utility from the load. Another way utility customers tamper with their utility meters is to put a shunt in the base of their meter to create a parallel electrical path that will not be monitored. Yet another common meter tampering method is to put one or more magnets on the meter. The magnets cause the meter to rotate slower than intended, resulting in a lower electric bill.
Utility customers also steal electricity from electrical utilities by tampering with the electrical lines leading into the property. Many utility customers bypass the meter within the meter housing by connecting a wire at the line side or input of the meter directly to the load side or output of the meter. In addition, some customers bypass their meters simply by tapping into an overhead power line on or near the property using a fish hook or similar device to bypass the meter. Other customers dig up underground power lines on their property and tap directly into those lines.
In any case, tampering with electric utility meters or power lines is dangerous and illegal. The traditional methods of detecting electricity theft include going to a customer's property to look for physical indications of tampering, gathering leads reported by the public, and investigating neighbors and relatives of customers found to be tampering to determine if they are also tampering. However, those methods are time consuming and expensive, so electric utilities developed methods of remotely detecting tampering. Several methods include monitoring meters for reverse flow events; power outages and blinks; load side voltages upon disconnecting power; magnetic detection using a Hall effect switch or a similar device; vibration or tilting of the meter; meter cover removal; and incorrect polyphase wiring. Further, transformers that feed primary meters electrically downstream therefrom may also be monitored so that the electricity or power readings at the transformer may be compared against the aggregated usage reported by the meters.
In addition to the above, substation feeder metering and advanced metering infrastructure (AMI) data may be incorporated into a power distribution model for the electric utility in order to determine the feeders with the greatest non-technical loss. Also, changes in current flow patterns may be detected before using thermal imaging to find overloaded transformers. Data analytics can be used to locate large spikes or drop offs from historical usage patterns at homes or commercial buildings. The data analytics can account for weather patterns, billing/payment information, comparisons to neighborhood consumption patterns, transformer to aggregate load comparisons, and various other factors.
While the above methods may be helpful in determining whether a customer is stealing electricity, those methods cannot perfectly determine whether electricity theft is occurring. None of the methods take advantage of every indicator of electricity theft. For example, none of the above methods monitor components downstream from the primary meter for indications of theft. In addition, the above methods only provide raw data that must be interpreted by the utility in order to determine how likely it is that electricity is being stolen by a customer. Interpretation of that raw data is time consuming and may not provide an indication of electricity theft or may provide a false positive indication of threat when viewed in isolation, depending on various factors. The utility may have to send someone to investigate, even if there is a low likelihood of theft.
It would therefore be desirable to provide a system and method for electricity theft detection that takes advantage of available electrical data in a power distribution system and that indicates the likelihood of electrical theft by an electric utility customer using multiple data sources.